{"api_version":2,"name":"piflow_lda_training","version":"1.0","role":"transform","type":"sdk","description":"Train a Spark online latent Dirichlet allocation model on Arrow records and emit the trained model as a ZIP byte stream.","icon":{"path":"M3 18h18M5 15l5-6 4 3 5-7"},"inputs":{"source":{"representations":["records"],"formats":[]}},"outputs":{"result":{"view":"bytes"}},"parameters":{"type":"object","properties":{"k":{"type":"integer","description":"Number of clusters or topics from 2 to 32.","minimum":2,"maximum":32},"maxIter":{"type":"integer","description":"Maximum optimizer iterations from 1 to 100.","minimum":1,"maximum":100}},"required":["k","maxIter"],"additionalProperties":false},"documentation":{"inputs":{"source":"Arrow records with equally sized numeric lists in features."},"outputs":{"result":"Native Spark ML model packaged as ZIP bytes. Connect to the matching prediction operator."},"parameters":{"k":{"description":"Number of clusters or topics from 2 to 32.","help":"Number of clusters or topics from 2 to 32.","example":2},"maxIter":{"description":"Maximum optimizer iterations from 1 to 100.","help":"Maximum optimizer iterations from 1 to 100.","example":8}},"example_job":"name: PiFlow online latent Dirichlet allocation training\nsteps:\n- id: read\n  operator: item_source@1.0\n  inputs: {}\n  params:\n    uri: context:/piflow/example.jsonl\n    view: records\n- id: process\n  operator: piflow_lda_training@1.0\n  inputs:\n    source: read\n  params:\n    k: 2\n    maxIter: 8\n- id: receive\n  operator: client_sink@1.0\n  inputs:\n    result: process\n"},"runtime":{"image":"sha256:edd926f5a144036f8fbfd7db6771067b44ef00023872fe68552d3419ab549e90","archive":{"sha256":"5b4a94b1342e29dcbdc310b0efa54739b4e221d84005420e9d54e305fdbd4546","size":425330122,"filename":"piflow-ecoflow-spark-runtime.tar.gz"}},"execution":{"executor":"jvm","entrypoint":"org.apache.spark.sql.piflow.LDATraining","config":{}},"source_public":true,"provenance":{"license":"Apache-2.0","source_repository":"https://github.com/cas-bigdatalab/piflow","source_revision":"777361c4e72b2c5457479683916b7c8e635289ff","build_instructions":"Compile src/ using piflow-ecoflow-spark@1.0. See job.yaml for a workflow and fixtures/example.jsonl for example records.","source_archive":{"sha256":"e01187c3e6691375453ee0c4a31ed421d9d5a38a8c684b6548883967f16edc47","size":11324,"filename":"piflow_lda_training-sources.zip"},"license_file":{"sha256":"c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4","size":11357,"filename":"LICENSE.txt"}},"artifact":{"sha256":"7a98abe012faa9d8b1794a41c47f63f9c3b642643d3c0ea50eef61c63158eb1d","size":21651,"filename":"operator.jar"},"categories":["PiFlow","Machine learning"],"trust":{"score":4,"assessed_by":"administrator","reviewed_at_ms":1795080080223},"publication":{"author":"dacp-examples","submitted_at_ms":1795080027220,"approved_at_ms":1795080080223,"last_submitted_at_ms":1795080027220,"last_approved_at_ms":1795080080223},"download_count":4}